Has Google fallen behind in the AI race?

Depends on which race. They seem to be on par or ahead of OpenAI in the research space: they're currently testing embodied knowledge in LLMs. [0]

However, it's doubtful if they were into the consumer AI space, since OpenAI more or less invented it by productizing their GPT.

[0] https://ai.googleblog.com/2023/03/palm-e-embodied-multimodal...

They are measurably behind since they have yet to put a LLM into a single product or even let anyone outside of the company try any of their models. Meanwhile, Microsoft has fully launched GPT in Bing and is driving ahead with deep office integration. GPT-4 is quite cool on its own, but their trick of prompt orchestration and allowing the LLM to interact with the office application APIs looks like it will have plenty of real world impact on users.

From the past week: https://youtu.be/Bf-dbS9CcRU

Google is in the danger zone.

I doubt it: they may be behind in the LLM space but LLM's aren't everything.

Don't forget Google has DeepMind: https://www.deepmind.com/. Its board/video game AI surely beats GPT, and its interactive agent / AlphaCode AI at least compares. They also have Imagen (https://imagen.research.google/) which is among the top image generators and one of the couple video generators out there.

Also, it's only been a few months since ChatGPT and the LLM explosion. They are surely doing research on LLMs, they just haven't released it as a consumer tool like OpenAI has.

Higher optical in-flight data-center information density using multiple all at once spectrum optical network physical layer gives Google a comparative and competitive edge. Maybe Google needs to find leadership teams that appreciate the power of A.I. and the information as they do in part but not whole have the edge on Apple or Microsoft. Not what you know but who you know in office i.t. data politicking wins the day so maybe the answer to A.I. can only be Microsoft even if Apple is better but the iPhone/Nexus/Pixel proved other wise and Microsoft failed in that department despite their enormous appetite for risk at failing to do-over where others have had cyclical success.

It is interesting that they have not tried to defend their patent indirectly based off of the Attention is all you need paper.

https://patents.google.com/patent/CA3050334A1

No.

There is more to 'AI' than LLMs, which is the current hype thing that everyone here is screaming about.

Seems like they are easily impressed by glorified bullshit generators hallucinating on basic questions than trustworthiness in AI.